Blackwell UltraLive Marketplace Asks Available

NVIDIA B300

B300 is the Blackwell Ultra GPU: 288 GB of HBM3e at 8 TB/s, fifth-generation Tensor Cores, and NVLink 5 at 1.8 TB/s. The extra 96 GB over B200 is what matters when a 200B-class model or a very long context does not fit in 192 GB. A typical node quote pairs the GPU with about 32 vCPUs and 184 GB of host RAM.

Browse Live Marketplace →All Hardware Fleet

Flexible Compute Modes

Deploy NVIDIA B300 Your Way: Docker Pods or Root KVM VPS

NovaPod lets you decide how you want to harness GPU power. Whether you need an instant pre-cached container stack, a full root virtual machine with dedicated public IP, or a multi-node NVLink cluster, pay only for what you run with second-by-second crypto billing.

Docker GPU PodCold Boot < 15s

Fast, Lightweight AI Containers

Pure GPU compute without hypervisor overhead. Boot straight into PyTorch, vLLM, ComfyUI, or JupyterLab.

1-second precision metering
37+ Pre-cached ML framework templates
Zero idle fee when instance is stopped

Deploy as Docker Pod

Root KVM VPSPersistent NVMe

Full Root Virtual Machine

Dedicated operating system (Ubuntu 24.04 LTS / Windows GPU Server) with complete sudo root SSH access from minute one.

Dedicated public IPv4 (all ports open)
Pause anytime; pay only $0.00014/GB/hr for disk
Run Docker-in-Docker, k3s, custom drivers

Deploy as Root VPS

Multi-GPU Cluster900 GB/s NVLink

Supercomputing & Distributed Training

Pool up to 8× NVIDIA B300 on a unified node fabric with Quantum-2 InfiniBand RDMA for multi-node scale.

Zero-loss all-reduce gradient exchange
PyTorch FSDP, DeepSpeed Zero-3 ready
Native GPUDirect Storage & RDMA fabric

Configure Cluster

Hardware Specifications

Architecture: NVIDIA Blackwell Ultra
VRAM: 288 GB HBM3e
Bandwidth: 8.0 TB/s
Tensor Cores: 5th generation
CUDA cores: 20,480 and up
FP4 dense: About 15 PFLOPS
FP8 dense: About 7,500 TFLOPS
FP32: About 120 TFLOPS
FP64: About 60 TFLOPS
NVLink: 1.8 TB/s
Reference host: 32 vCPUs, 184 GB RAM
Local disk: On the order of 250 GB NVMe, varies by offer
8-GPU pool: About 2.3 TB HBM3e
TDP: About 1,200 W
Form factor: SXM6

The GPU itself

B300 is the Blackwell Ultra chip: 288 GB of HBM3e at 8 TB/s, fifth-generation Tensor Cores, NVLink 5 at 1.8 TB/s, and a board power around 1,200 W. CUDA core counts published for the part sit above 20,000. Dense FP4 is about 15 PFLOPS and dense FP8 about 7,500 TFLOPS. FP64 is about 60 TFLOPS and FP32 about 120 TFLOPS, which tells you this is an AI card, not a double-precision HPC card.

The menu row assumes a single-GPU host of about 32 vCPUs and 184 GB of RAM, with on the order of 250 GB of local NVMe. An 8-GPU HGX node pools about 2.3 TB of HBM3e. That node is the unit most training jobs actually buy.

What 288 GB changes

Bandwidth matches B200 at 8 TB/s. The reason to pay for B300 is capacity: 50 percent more memory than 192 GB. A 200B-class dense model in FP8 can sit on one GPU. A trillion-parameter mixture-of-experts can be planned across one 8-GPU node instead of two nodes that exist only to hold weights. Context windows past several hundred thousand tokens stop evicting the KV cache as early.

If the model and the cache already fit in 192 GB, the extra 96 GB is idle silicon. Take B200. If you do not need FP4 at all, H200 at 141 GB is the Hopper card with a lower ask.

How people run it

Training containers are the current NVIDIA PyTorch and NeMo images with Transformer Engine new enough for FP8, and FP4 only if you have checked that recipe. Serving is vLLM or TensorRT-LLM with the quantization the model was exported for. Eight-way jobs use tensor and pipeline parallel inside the node before they touch the network.

NVLink 5 at 1.8 TB/s is the in-node fabric: full NVSwitch across eight GPUs, GPU-to-GPU traffic that does not bounce through the CPU, and the usual NCCL collectives. That is what makes an 8-GPU box behave like one machine. Past eight GPUs you are on the GB300 rack or on InfiniBand between nodes.

Billing

This marketplace does not publish a standing B300 rate. When a host lists one, the card shows the hourly ask and we bill that ask per second. A dedicated offer stays until you stop it. A reclaimable offer is only for work you can checkpoint.

Buying the card is a different problem. List prices for a single B300 are in the tens of thousands of dollars, and an 8-way node with networking and cooling is an order of magnitude more. Renting wins unless the GPU would be busy most of the year and you already operate the rack.

What the Numbers Mean in Production

Memory vs H100: 3.6×
Memory vs B200: +96 GB
FP4 dense: ~15 PFLOPS
8-GPU pool: ~2.3 TB

Storage & Tooling Infrastructure

High-Speed Storage, Native Ports & Crypto Rails

Every NVIDIA B300 instance on NovaPod connects directly to high-throughput infrastructure designed to eliminate I/O bottlenecks and onboarding delays.

25 Gb/s Shared Network Storage

Mount large model checkpoints and training sets across one or multiple instances simultaneously. Data outlives single pods and stays accessible when instances stop.

JupyterLab, VS Code & SSH

Connect from your local terminal with standard SSH keys or open web-based IDEs directly in browser tabs without managing complex reverse proxy configs.

Non-Custodial Multi-Chain Rails

Deposit USDT, USDC, SOL, ETH, or BNB to your dedicated personal deposit address. Confirmed on-chain in seconds with zero mandatory identity verification.

Target Workloads for NVIDIA B300

Frontier training

288 GB per GPU lets a larger dense or mixture-of-experts model sit on fewer cards than a 192 GB plan.

200B and up, denseLong-context transformersMultimodal pre-trainingOne 8-GPU node instead of two

Large-model serving

FP4 plus the extra 96 GB cuts tensor-parallel splits for models that spill off a B200.

200B and larger inferenceVery long contextsAgent traces with a large KV cacheHigh concurrency, not a single stream

Research that is memory-bound

Use it when the profile says the run died on capacity. If a cheaper Hopper card already fits, this GPU is idle silicon.

Architecture searchesLong in-context runsLarge replay buffersCheckpoints that must stay on GPU

Inside one NVLink node

Eight B300s with NVLink 5 behave like one machine. Past eight GPUs you are on the GB300 rack or on a network between nodes.

Tensor and pipeline parallelNCCL collectivesGPU-to-GPU, not through the CPUQuote a rack only after the node is full

When to Choose NVIDIA B300

Pick B300 when 192 GB is the wall

One B300 can replace two B200s that existed only to hold weights.

Pick B200 when the model fits

Bandwidth stays near 8 TB/s. You would be paying for 96 GB you do not touch.

Pick H200 when you do not need FP4

141 GB of HBM3e on Hopper is the cheaper long-context card for jobs that already fit.

Pick GB300 when eight GPUs are not enough

The rack is the same silicon with Grace and a shared NVLink domain.

When 288 GB changes the job

Use B300 when weights, optimizer state, or a 1M-token KV cache do not fit in 192 GB. If they fit, B200 is the cheaper Blackwell card. The bandwidth does not jump versus B200; the capacity does.

Software

Current PyTorch, TensorFlow, and JAX containers with CUDA 12.5-class libraries, cuDNN, and TensorRT are what you want. FP4 needs a recent Transformer Engine. Bring that image on deploy.

Frequently Asked Questions

What changed versus B200?

VRAM goes from 192 GB to 288 GB. Bandwidth stays near 8 TB/s. Tensor throughput and board power are higher. The purchase is capacity, not a faster bus.

Does 288 GB change inference?

Yes when the weights or the KV cache do not fit in 192 GB. If they do, B200 is usually the better rate.

Is one B300 cheaper than two B200s?

Often, if the second B200 existed only to hold memory. If the model already fits on one B200, pay for the B200.

Can I use it for inference only?

Yes. Under about 100B parameters, H100 or B200 usually win on dollars per token. B300 is for models and contexts that spill.

What host does the menu assume?

A single-GPU shape around 32 vCPUs, 184 GB of RAM, and 288 GB of VRAM. The live offer is the contract. Clusters are a separate conversation.

Which software?

A current NVIDIA PyTorch or NeMo image, CUDA 12.5-class libraries, and a Transformer Engine new enough for the precision you picked. FP4 needs that recent engine. Training that has not been checked stays on FP8 or BF16.

What is NVLink 5 doing?

1.8 TB/s between GPUs in the node, with NVSwitch across eight cards, so collectives do not bounce through the CPU. That is what makes the 2.3 TB pool usable.

Should I buy the card?

A single B300 is a five-figure part, and an 8-way node with power and networking is far more. Rent unless the GPU would be busy most of the year and you already operate the rack.

How is it billed here?

Per second from the live ask. Stop the machine and the charge stops. There is no separate monthly B300 plan.

Alternative Hardware Options

Ready to Deploy NVIDIA B300?

Launch an instance on the live marketplace in under 15 seconds. Fund your balance with cryptocurrency and pay per second with zero commitments.

Launch on Marketplace →